Structural Enforcement of Goal Integrity in AI Agents via Separation-of-Powers Architecture

arXiv:2604.23646v1 Announce Type: new Abstract: Recent evidence suggests that frontier AI systems can exhibit agentic misalignment, generating and executing harmful actions derived from internally constructed goals, even without explicit user requests. Existing mitigation methods, such as Reinforcement Learning from Human Feedback (RLHF) and constitutional prompting, operate primarily at the model level and provide only probabilistic safety guarantees. We propose the Policy-Execution-Authorization (PEA) architecture, a "separation-of-powers" design that enforces safety at the system level. PEA decouples intent generation, authorization, and execution into independent, isolated layers connected via cryptographically constrained capability tokens. We present five core contributions: (C1) an Intent Verification Layer (IVL) for ensuring capability-intent consistency; (C2) Intent Lineage Tracking (ILT), which binds all executable intents to the originating user request via cryptographic anchors; (C3) Goal Drift Detection, which rejects semantically divergent intents below a configurable threshold; (C4) an Output Semantic Gate (OSG) that detects implicit coercion using a structured K imes I imes P threat calculus (Knowledge, Influence, Policy); and (C5) a formal verification framework proving that goal integrity is maintained even under adversarial model compromise. By shifting agent alignment from a behavioral property to a structurally enforced system constraint, PEA provides a robust foundation for the governance of autonomous agents.

Related

• Atomic Decision Boundaries: A Structural Requirement for Guaranteeing Execution-Time Admissibility in Autonomous Systems
• Formalizing the Safety, Security, and Functional Properties of Agentic AI Systems
• Discovering Agentic Safety Specifications from 1-Bit Danger Signals
• AI Organizations are More Effective but Less Aligned than Individual Agents

Source: arXiv cs.AI | 2026-04-28